Research map / Video / Video generation
GRE-Diff: Gaussian Room Embeddings for Structured Layout Diffusion
Jing Wang, Haoran Xiong, Zihao Yan, Minglun Gong, Hui Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add global attention to GuidanceNet (with local attention) (ablation) | GRE-Diff · RPLAN generation | FID: 4.57 -> 4.36 | Improved |
| add local attention to GuidanceNet (with global attention) (ablation) | GRE-Diff · RPLAN generation | FID: 4.49 -> 4.36 | Improved |
| add local attention to GuidanceNet (with global attention) (ablation) | GRE-Diff · RPLAN generation | COV: 86.71% -> 96.09% | Improved |
| add global attention to GuidanceNet (with local attention) (ablation) | GRE-Diff · RPLAN generation | COV: 86.52% -> 96.09% | Improved |
| add global attention to GuidanceNet (with local attention) (ablation) | GRE-Diff · RPLAN generation | MMD: 0.153 -> 0.107 | Improved |
| add local attention to GuidanceNet (with global attention) (ablation) | GRE-Diff · RPLAN generation | KID (x10^-3): 0.82 -> 0.96 | Hurt |
| add local attention to GuidanceNet (with global attention) (ablation) | GRE-Diff · RPLAN generation | F1 BC-RC: 98.23% -> 99.21% | Improved |
| use GRE inside diffusion dynamics rather than only as external condition (ablation) | GRE-Diff · RPLAN generation | quality | Improved |
| remove Graph2Plan's retrieval-based refinement (diagnostic) | GRE-Diff · RPLAN generation (Graph2Plan) | FID: 4.81 -> 20.10 | Hurt |
| use GRE-Diff instead of ChatHouseDiffusion for Add editing | GRE-Diff · RPLAN editing (Add) | F1 BC-RC: 68.45 -> 86.27 | Improved |
| use GRE-Diff instead of ChatHouseDiffusion for Delete editing | GRE-Diff · RPLAN editing (Delete) | F1 BC-RC: 58.26 -> 98.58 | Improved |
| use GRE-Diff instead of ChatHouseDiffusion for Anchor & Move editing | GRE-Diff · RPLAN editing (Anchor & Move) | F1 BC-RC: 73.06 -> 97.59 | Improved |
Closest papers by meaning
- OptiScene: LLM-driven Indoor Scene Layout Generation via Scaled Human-aligned Data Synthesis and Multi-Stage Preference Optimization (NeurIPS 2025 · similarity 0.89)
- MaskPLAN: Masked Generative Layout Planning from Partial Input (CVPR 2024 · similarity 0.87)
- GSDiff: Synthesizing Vector Floorplans via Geometry-enhanced Structural Graph Generation (AAAI 2025 · similarity 0.86)
- Tokenization Allows Multimodal Large Language Models to Understand, Generate and Edit Architectural Floor Plans (CVPR 2026 · similarity 0.86)
- Direct Numerical Layout Generation for 3D Indoor Scene Synthesis via Spatial Reasoning (NeurIPS 2025 · similarity 0.86)
- DeBaRA: Denoising-Based 3D Room Arrangement Generation (NeurIPS 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
